The molecular structure of selected South African coal-chars to elucidate fundamental principles of coal gasification

(1)

The molecular structure of selected South

African coal-chars to elucidate

fundamental principles of coal

gasification

M.J. Roberts

22061452

Thesis submitted for the degree

Doctor Philosophiae

in

Chemical Engineering

at the Potchefstroom Campus of the

North-West University

Promoter:

Prof R.C. Everson (North-West University)

Co-promoter:

Dr G. Domazetis MD Clean Coal Technology

Pty Ltd (La Trobe University, Australia)

(2)

i

Declaration

I, Mokone J. Roberts, hereby declare that a thesis with the title:

THEMOLECULARSTRUCTUREOFSELECTEDSOUTHAFRICANCOAL-CHARSTO ELUCIDATEFUNDAMENTALPRINCIPLESOFCOALGASIFICATION

is my own work, except where acknowledged, and has not been submitted at any other university either in whole or in part.

Signed at Potchefstroom on the ………... day of October 2015

__________________________

(3)

ii

Acknowledgements

First and foremost I acknowledge the sustaining power of faith in God Almighty during the times of uncertainty and testing I experienced in this thesis.

I would like to thank the North-West University, Potchefstroom Campus, for matching me with my PhD study leader and promoter, Professor Raymond Everson, under the directorship of Professor Frans Waanders, for so patiently allowing me an intellectual freedom in the most challenging research I ever experienced. I felt extremely lucky to have Professor Raymond Everson who walked with me and cared so much about my work every step of the way. His calm but enthusiastic and smart demeanour steered my mental abilities to such answers, comprehension and rationale I never thought I could develop. So closely alongside him I would also like to thank my assistant promoter, Professor Hein Neomagus, whose energetic and probing mind would run me around a point until I either captured or knocked it down. A research group meeting would always feel strange and somewhat incomplete if neither of these two Professors were present. The involvement of all my co-promoters was indispensable since the inception of this project. Professor Jonathan Mathews offered me a remote contract from the Penn State University with his curiosity on the atomistic models of coal and dedication to the coal science and technology. Together with one of his protégés, Dr Daniel van Niekerk from SASOL, their fine expertise is gratefully acknowledged. I appreciate and thank the efficient participation of Dr George Domazetis who, despite using different time-zones and programs from as far as the La Trobe University, Melbourne, shared his abundant experiences with the atomistic coal science that involved simulations in both the semi-empirical and high-accuracy density functional theory techniques. I express my gratitude to Dr Cornie Van Sittert, who facilitated the core resources of my thesis throughout in respect with the laboratory for molecular modelling at North-West University. In different combinations, these promoters contributed to the successful publications described chapters 3-5.

(4)

iii I appreciate the generous collaboration, growing and impressive skills of one of my PhD colleagues, Gregory N. Okolo, with the coal characterisation and experimental work related to coal gasification, including his ability and speed to acquire and interpret results. Okolo’s efforts nearly touched all the aspects of my thesis, in particular, the two publications (Chapters 3 and 5), including Chapter 6. Okolo made a phrase I often heard in the streets of Lagos, Nigeria, during my brief visits in the mid-2000s: “I am we”, to appear so glaringly in my experiences and memories. The contribution of other junior students like Marie-Louise Visser, Thabiso Moropa, Lebohang Mgano and Thandeka Chaldien in their different years of studies at the North-West University is highly acknowledged. Either in single or joint efforts, high-quality samples for different analyses were produced, at times on 24/7 turnarounds. Coupled with their preparedness to cooperate with other laboratory users, including their ability to acquire support and technical assistance from the Laboratory Manager, the two main engineering technicians and their assistants, these students earned themselves respect among their peers and were empowered with some of the best experiences and practices in coal sample treatment for research purposes.

Most results in this thesis were made possible with support of specialists and sponsors from different divisions, institutions and organisations. In no specific order, I thank SANEDI and ESKOM for supporting this thesis; Mr Aaron Mulaudzi of the Exarro Resources’ Grootegeluk Coal Mine and Mr J.P. Bergh from the Anglo American Corporation’s Goedehoop Coal Mine for their dedicated effort to identify and supply me with the desired coal samples used in this thesis; Professor Manie Vosloo of the Chemical Resource Beneficiation faculty of the North-West University for his assistance with the initial layout of my molecular modelling requirements; Professors Rosemary Falcon and Nicola Wagner from the University of the Witwatersrand, Johannesburg, including Ms Vivien Du Cann from Petrographic SA, for offering their resources, guidelines and interesting discussion on coal petrography, which formed part of Chapter 3; Mr Mohamed Jaffer under the instructions of Professor Trevor Sewell at the University of Cape Town for going all the way outside their usual material analyses to

(5)

iv provide me with the high resolution transmission electron microscopy images of my coal char samples that culminated in chapters 4 and 5; the University of Nottingham’s Professor Colin Snape and the Stellenbosch University’s Ms Heidi Assumption and Dr Susanne Causemann for their generosity to devote their resources and knowledge of the solid-state nuclear magnetic resonance for my coal chars; the involvement and such steering questions of Professor Jenny Jones from the University of Leeds, UK, which opened my wider understanding of the computational chemistry.

I express gratitude to all of the North-West University faculty members for their help and support towards the accomplishment of my thesis. Professors John Bunt, Quentin Campbell and Christien Strydom, including Dr Rufaro Kaitano strengthened me during my lows and desperations in small but best ways they could despite their busy schedule. I also place on record, my sense of gratitude to everyone else in the campus for having directly or indirectly lent their hand in the daily and final developments of my thesis, including past and present students and research administrators.

I deeply thank my spiritual mother, MaKhabele Elizabeth Mpheqeke, for her unconditional trust in me her devoted responsibility based on her exceptional faith in God, which influenced my voyage enormously when I got troubled and weary. My brothers and sisters from all houses, as well as my in-laws have been so generous with their love and understanding despite extreme differences in our levels of literacy.

Lastly, and most importantly, I wish to thank with abundance of love and warmth my wife MaRoberts and daughters Tiffy, Ratwe, Bontle and Thori Roberts. I shall forever recognise their sacrifice to allow me to pursue these most difficult full-time studies far from home.

I dedicate this thesis to my parents who did not leave long enough see the first PhD graduate in the family. In so much humble ways, yet against all odds, their love raised me up and gave me strength and values of a winner and survivor I am hitherto.

(6)

v

Abstract

Advances in the knowledge of chemical structure of coal and development of high performance computational techniques led to more than hundred and thirty four proposed molecular level representations (models) of coal between 1942 and 2010. These models were virtually on the carboniferous coals from the northern hemisphere. There are only two molecular models based on the inertinite- and vitrinite-rich coals from the southern hemisphere. The current investigation is based on the chars derived from the Permian-aged coals in two major South African coalfields, Witbank #4 seam and Waterberg Upper Ecca. The two coals were upgraded to 85 and 93% inertinite- and vitrinite-rich concentrates, on visible mineral matter free basis. The coals were slow heated in inert atmosphere at 20 ℃ min-1 to 450, 700 and 1000 ℃ and held at that temperature for an hour. After the HCl-HF treatment technique at ambient temperatures, the characteristics of the coals and chars were examined with proximate, ultimate, helium density, porosity, surface area, petrographic, solid-state 13C NMR, XRD and HRTEM analytical techniques. The results largely showed that substantial transitions occurred at 700-1000 ℃, where the chars became physically different but chemically similar. Consequently, the chars at the highest temperature (1000 ℃) drew attention to the detailed study of the atomistic properties that may give rise to different reactivity behaviours with CO2 gas.

The H/C atomic ratios for the inertinite- and vitrinite-rich chars were respectively 0.31 and 0.49 at 450 ℃ and 0.10 and 0.12 at 1000 ℃. The true density was respectively 1.48 and 1.38 g.cm-3 at 450 ℃ and 1.87 and 1.81 g.cm-3 at 1000 ℃. The char form results from the petrographic analysis technique indicated that the 700-1000 ℃ inertinite-rich chars have lower proportions of thick-walled isotropic coke derived from pure vitrinites (5-8%) compared with the vitrinite-rich chars (91-95%). This property leads to the creation of pores and increases of volume and surface area as the softening walls expand. It was found that the average crystallite diameter, La, and the mean length of

the aromatic carbon fringes from the XRD and HRTEM techniques, respectively, were in good agreement and made a definite distinction between the 1000 ℃ inertinite- and

(7)

vi vitrinite-rich chars. The crystallite diameter on peak (10) approximations, La(10), of

37.6Å for the 1000 ℃ inertinite-rich chars fell within the HRTEM’s range of minimum-maximum length boundary of 11x11 aromatic fringes (27-45Å). The La (10) of 30.7Å for

the vitrinite-rich chars fell nearly on the minimum-maximum length range of 7x7 aromatic fringes (17-28Å.) The HRTEM results showed that the 1000 ℃ inertinite-rich chars comprised a higher distribution of larger aromatic fringes (11x11 parallelogram catenations) compared with a higher distribution of smaller aromatic fringes (7x7 parallelogram catenations).

The mechanism for the similarity between the 700-1000 ℃ inertinite- and vitrinite-rich chars was the greater transition occurring in the vitrinite-rich coal to match the more resistant inertinite-rich coal. This emphasised that the transitions in the properties of vitrinite-rich coals were more thermally accelerated than those of the inertinite-rich coals. The similarity between the inertinite- and vitrinite-rich chars was shown by the total maceral reflectance, proximate, ultimate, skeletal density and aromaticity results. Evidence for this was the carbon content by mass for the inertinite- and vitrinite-rich chars of respectively 90.5 and 85.3% at 450 ℃ and 95.9 and 94.1% at 1000 ℃. The aromaticity from the XRD technique was respectively 87 and 77% at 450 ℃ and 98 and 96% at 1000 ℃. A similar pattern was found in the hydrogen and oxygen contents, the atomic O/C ratios and the aromaticity from the NMR technique.

The subsequent construction of large-scale molecular structures for the 1000 ℃ inertinite-rich chars comprised 106 molecules constructed from a total of 42929 atoms, while the vitrinite-rich char model was made up of 185 molecules consisting of a total of 44315 atoms. The difference between the number of molecules was due to the inertinite-rich char model comprising a higher distribution of larger molecules compared with the vitrinite-rich char model, in agreement with the XRD and HRTEM results. These char structures were used to examine the behaviour on the basis of gasification reactivity with CO2.

The density functional theory (DFT) was used to evaluate the interactions between CO2

(8)

vitrinite-vii rich South African coals. The construction of char models used the modal aromatic fringes (fringes of highest frequencies in size distributions) from the HRTEM, for the inertinite- and vitrinite-rich chars, respectively (11x11 and 7x7 parallelogram-shaped aromatic carbon rings). The structures were DFT geometrically optimized and used to measure reactivity with the Fukui function, f+(r) and to depict a representative reactive carbon edge for the simulations of coal gasification reaction mechanism with CO2 gas.

The f+(r) reactivity indices of the reactive edge follows the sequence: zigzag C remote from the tip C (Czi = 0.266) > first armchair C (Cr1 = 0.087) > tip C (Ct = 0.075) > second

armchair C (Cr2 = 0.029) > zigzag C proximate to the tip C (Cz = 0.027). The DFT

simulated mean activation energy, ∆Eb, for the gasification reaction mechanism

(formation of second CO gas molecule) was 233 kJ mol-1. The reaction for the formation of second CO molecule is defines gasification in essence. The experimental activation energy determined with the TGA and random pore model to account essentially for the pore variation in addition to the gasification chemical reaction were found to be very similar: 191 ± 25 kJ mol-1 and 210 ± 8 kJ mol-1; and in good agreement with the atomistic results. The investigation gave promise towards the utility of molecular representations of coal char within the context of fundamental coal gasification reaction mechanism with CO2.

(9)

viii

Opsomming

Vooruitgang in die kennis van die chemiese struktuur van steenkool en die ontwikkeling van hoëwerkverrigting-rekenaartegnieke het gelei meer as honderd vier en dertig voorgestelde molekulêre vlak voorstellings (modelle) van steenkool tussen 1942 en 2010. Hierdie modelle was feitlik almal steenkool van die noordelike halfrond. Daar is net twee molekulêre modelle gebaseer op die inertiniet- en vitrinietryke steenkool uit die suidelike halfrond. Die huidige ondersoek is gebaseer op die sintels afgelei van die Perm-ouderdom van steenkool in twee groot Suid-Afrikaanse steenkoolvelde, Witbank # 4 rif en Waterberg-Bo-Ecca. Die twee steenkole is opgegradeer tot 85 en 93% inertiniet- en vitrinietryke konsentrate, respektiewelik, op 'n sigbare mineraal-vry basis. Die steenkool is stadig verhit in 'n inerte atmosfeer teen 20°C min-1_{tot 450, 700 en 1000°C}

en by daardie temperatuur vir een uur gehou. Na toepassing van die HCl/HF-behandelingstegniek by omgewingstemperatuur, is die eienskappe van die steenkool en sintels ondersoek met kort-analise, eind-analise, heliumdigtheid, porositeit, oppervlakte, petrografiese, vastetoestand13C KMR-, XSD en HRTEM analitiese tegnieke. Die resultate het getoon dat aansienlike oorgange plaasgevind het by 700-1000°C, waar die sintels fisies verander is, maar chemies soortgelyk was. Gevolglik het die sintels op die hoogste temperatuur (1000°C) die aandag gelei na die gedetailleerde studie van die atomiese eienskappe wat aanleiding gee tot verskillende reaktiwiteit-gedrag met CO2

gas.

Die H/C atoomverhoudings vir die inertiniet- en vitrinietryke sintels was onderskeidelik 0.31 en 0.49 by 450°C en 0.10 en 0.12 by 1000°C. Die ware digtheid was onderskeidelik 1.48 en 1.38 g cm-3 by 450°C en 1.87 en 1.81 g cm-3 by 1000°C. Die sintelvorm-resultate van die petrografiese analisetegniek het aangedui dat die 700-1000 °C inertinietryke sintels 'n laer verhouding van dikwandige isotropiese kooks het, afgelei van suiwer vitriniete (5-8%), in vergelyking met die vitrinietryke sintels (91-95%). Hierdie eienskap lei tot die skepping van porieë en verhoog die volume en oppervlakte soos die versagte wande uitsit. Daar is gevind dat die gemiddelde kristallietdeursnee en die gemiddelde lengte van die aromatiese koolstofrande van die XSD- en HRTEM-tegnieke,

(10)

ix onderskeidelik, goed ooreengekom het en 'n definitiewe onderskeid tussen die 1000°C inertiniet- en vitrinietryke sintels getref het. Die La (10) van 37.6Å vir die 1000°C

inertinietryke sintels het binne die HRTEM reeks van minimum-maksimum lengtegrens van 11x11 aromatiese rande (27-45Å) geval. Die La (10) van 30.7Å vir die vitrinietryke

sintels het byna op die minimum-maksimum lengte reeks 7x7 aromatiese rand (17-28Å) geval. Die HRTEM-resultate het getoon dat die 1000 °C inertinietryke sintels bestaan het uit 'n hoër verspreiding van groter aromatiese rande (11x11 parallelogram-aaneenskakelings) in vergelyking met 'n hoër verspreiding van kleiner aromatiese rande (7x7 parallelogram-aaneenskakelings).

Die meganisme vir die ooreenkoms tussen die 700-1000 °C inertiniet- en vitrinietryke sintels was die groter oorgang in die vitrinietryke steenkool om die meer bestande inertiniet-ryke steenkool te ewenaar. Dit beklemtoon dat die oorgange in die eienskappe van vitrinietryke steenkool meer termies geaktiveer is as dié van die inertinietryke steenkool. Die ooreenkoms tussen die inertiniet- en vitrinietryke sintels is getoon deur die resultate van totale maseraalreflektansie, kort-analise, eind-analise, geraamde digtheid en aromatisiteit. Bewyse hiervoor was die koolstofinhoud volgens massa vir die inertiniet- en vitrinietryke sintels van onderskeidelik 90.5 en 85.3% by 450 °C en 95.9 en 94.1% by 1000 °C. Die aromatisiteit volgens die XSD-tegniek was onderskeidelik, 87 en 77% by 450°C en 98 en 96% teen 1000°C. 'n Soortgelyke patroon is vir die waterstof- en suurstofinhoud gevind, die atoom O/C verhoudings en die aromatisiteit dmv die KMR-tegniek.

Die daaropvolgende konstruksie van grootskaalse molekulêre strukture vir die 1000°C inertinietryke sintels bestaan uit 106 molekules wat saamgestel is uit 'n totaal van 42 929 atome, terwyl die vitrinietryke sintelmodel opgemaak is uit 185 molekules wat bestaan uit 'n totaal van 44 315 atome. Die verskil tussen die aantal molekules was te wyte aan die inertinietryke sintelmodel wat bestaan uit 'n groter verspreiding van groter molekules in vergelyking met die vitrinietryke sintelmodel, in ooreenstemming met die XSD- en HRTEM-resultate. Hierdie sintelstrukture is gebruik om die gedrag op grond van vergassingsreaktiwiteit met CO2 te ondersoek.

(11)

x Die DFT is gebruik om die interaksies tussen CO2 en die atomiese voorstellings van

steenkool sintel, afgelei van die inertiniet- en vitrinietryke Suid-Afrikaanse steenkool, te evalueer. Die konstruksie van sintelmodelle het die aromatiese rande van die HRTEM gebruik, van die hoogste frekwensies vir die inertiniet- en vitrinietryke sintels, onderskeidelik (11x11 en 7x7). Die strukture is DFT-meetkundig geoptimaliseer en gebruik om reaktiwiteit met die Fukui-funksie, f+(r) te meet. Die 3x3 aromatiese steenkoolsintelstrukture is gebruik om 'n verteenwoordigende reaktiewe koolstofrand vir die evaluering van die steenkoolvergassingsreaksiemeganisme met CO2 gas uit te

beeld. Die f+(r) reaktiwiteitsindekse van die reaktiewe rand het die volgorde: sigsag C ver van die punt C (Czi = 0.266)> eerste stoel C (Cr1 = 0.087) > punt C (Ct = 0.075)>

tweede stoel C (Cr2 = 0.029)> sigsag C naby die punt C (Cz = 0.027). Die

DFT-gesimuleerde gemiddelde aktiveringsenergie op die reaktiewe rand was 233 kJ mol-1. Die aktiveringsenergie Ea, bepaal met behulp van die ewekansige poriemodel om die

porievariasie bykomend tot die chemiese reaksie in ag te neem, is ook gevind baie soortgelyk te wees vir die ontaste vitriniet- (191 ± 25 kJ mol-1) en inertinietryke sintels (210 ± 8 kJ mol-1). Goeie ooreenkoms is gevind tussen die DFT- en eksperimentele resultate. Die ondersoek hou belofte in vir die nut van molekulêre voorstellings van steenkoolsintel binne die konteks van die fundamentele steenkoolvergassingsreaksiemeganisme met CO2.

(12)

xi

Keywords

 Advanced coal properties

 Pyrolysis

 Micro-image processing and analyses

 High performance computing

 Computational chemistry

 Molecular modelling

 Density functional theory

 Energy

 Reactivity

 Experimental verification

 Inertinite- and vitrinite-rich chars

 Fukui function

(13)

xii

List of figures

Figure 2-1. The XRD spectra showing the three characteristic peaks (Lu et al., 2001:1821; Everson et al., 2013:148). ... 31 Figure 2-2. Determination of amorphous fraction of carbon (xA), from Gaussian curve

deconvolution (Everson et al., 2013:148). ... 33 Figure 2-3. The conventional ((a)-(f) up to (i)) and the advanced filtration technique ((a)-(h)

up to (j)-(k)) (Sharma et al., 1999:1203). ... 36 Figure 2-4. Typical HRTEM microphotograph (left-hand side) with its lattice fringes

(right-hand side), after the advanced filtration technique (Sharma et al.,

2002:54; Sharma et al., 1999:1203). ... 36 Figure 2-5. The molecular representations of bituminous coal (Mathews and Chaffee,

2012:1): (a) Fuchs and Sandoff structure (Fuchs and Sandhoff,

1942:567), (b) Given (Given, 1960:147), (c) adapted from Given (Given, 1962:39), (d) Meyers (Meyers, 1981), (e) Cartz and Hirsch (Cartz and Hirsch, 1960:557), (f) Ladner (as printed in Gibson (Mathews and Chaffee, 2012:1)) (Gibson, 1978:67), (g) Solomon (Solomon, 1981:61), (h) Hill and Lyon (Hill and Lyon, 1962:36), (i) adapted from Wiser (Wiser, 1984:325), (j) Shinn (Shinn, 1984:1187). Structures are reprinted with

permission of the copyright holders (Mathews and Chaffee, 2012:1). ... 43 Figure 2-6. Selected molecular representations of bituminous coal (Mathews and Chaffee,

2012:1): (a-f) the space-filling representation of Spiro (Spiro,

1981:1121), (a) Wiser model, (b) Given model, (c) Solomon model, (d) Wiser model, (e) Solomon model, (f) Solomon model globular

configuration, (g) Lazarov and Marinov (Lazarov and Marinov, 1987:411), (h) Zao Zhuang coal model (Nomura, Artok, Murata, Yamamoto, Hama, Gao, and Kidena) (Nomura et al., 1998:512), (i) Takanohashi et al. Upper Freeport model (Takanohashi and

Kawashima, 2002:379), (j) Narkiewicz and Mathews (Narkiewicz and Mathews, 2008:3104), (k-l) inertinite-rich and vitrinite-rich models of Van

(24)

xxiii

Niekerk and Mathews (Niekerk and Mathews, 2010:73). Structures were reprinted with permission of the copyright holders (Mathews and

Chaffee, 2012:1). ... 45 Figure 2-7. Structure of coal-to-char transitions derived via molecular modelling

approaches (Mathews et al., 2011:718). ((a), left-hand side) drop-tube reactor temperature profile and SEM of coal and char particles, and molecular models of coal and chars for Lewiston–Stockton vitrinite (Mathews et al., 1998:136). ((b), right-hand side) Pittsburgh #8 coal to char transitions ((b), right hand side) for wire-mesh generated chars (Jones et al., 1999:1737). Copyright Elsevier (1999) (Mathews et al.,

2011:718). Reprinted with permission (Mathews et al., 2011:718). ... 46 Figure 2-8. Coal char structures. ((a), left-hand side) Marzec oligomeric C173H88O2 model

for low-temperature coal char (Marzec, 1997:837). In ((a), left-hand side), arrows indicate H atoms that are less than 0.1 nm apart; strong repulsive forces between them make the planar conformer unstable (Marzec, 1997:837). ((b), right-hand side) Castro-Marcano Illinois No. 6 coal chars; composed of 7458 atoms (C5743H1511O131N61S12) within 66

molecules (Castro-Marcano et al., 2012:1272), where green = C, white = H, red = O, blue = N and yellow = S. ... 47 Figure 2-9. Illustration of four different CO2 chemisorption orientations on different char

models. Zigzag models (z4, z5, z7, z9), armchair models (a3, a5) and tip models (t3, t4). Numbers after the letters represent the number of

six-membered carbon rings (Montoya et al., 2003:29). ... 58 Figure 3-1. Schematic representation of char generation equipment. ... 80 Figure 3-2. Surface area and porosity transitions of the IR and VR particles with

temperature. ... 88 Figure 3-3. Total (Mean random) reflectance scan (Rsc%) for coals and chars... 92 Figure 3-4. Maceral contents for (a) inertinite-rich coals and chars and (b) vitrinite-rich

(25)

xxiv

Figure 3-5. Transformation of inertinites, vitrinites and inerts from (a) coals (left-hand side) to (b) char forms (right-hand side) with increasing temperature, where IR and VR = inertinite- and vitrinite-rich, respectively. ... 95 Figure 3-6. The NMR spectra, where: (a) - (d) = inertinite-rich particles (chars-coal) and

(e) – (h) = vitrinite-rich particles (chars - coal); @numbers =

temperatures in ℃... 97 Figure 3-7. Baseline corrected and smoothed XRD diffractograms of coals and chars,

where: (a) and (b) = inertinite- and vitrinite-rich particles, respectively; (c)–(f) = inertinite- vs vitrinite-rich particles. Numbers in the legends =

temperatures in ℃... 99 Figure 4-1. Dissections showing cropped HRTEM images. ... 118 Figure 4-2. The 13C NMR spectra of the inertinite- and vitrinite-rich chars generated at

1000 ℃ ((a) and (c), left-hand side) and their corresponding parent coals ((b) and (d), right-hand side). ... 123 Figure 4-3. Baseline corrected and smoothened XRD diffractograms of chars generated at

1000 ℃ ((a), left-hand side) and parent coals ((b), right-hand side). ... 125 Figure 4-4. HRTEM results. HRTEM results. Cropped image ((a), left-hand side) and

skeletonised image after image processing ((b), right-hand side). ... 127 Figure 4-5. A 21x21 parallelogram-shaped aromatic fringe showing maximum (MaxL) and

minimum (MinL) lengths. ... 128 Figure 4-6. Aromatic fringe length distributions from the HRTEM. ... 131 Figure 4-7. An example of 25 aromatic fringes of length distributions from the HRTEM,

used during the model constructions. ... 132 Figure 4-8. Example of geometric representations of char molecules after trimming an

(26)

xxv

Figure 4-9. Cross-linked ((a), left-hand side) and individual ((b), right-hand) molecules with their functional atoms. Green = C and white = H. ... 134 Figure 4-10. Geometry optimised and annealed 3D molecular models of inertinite-rich

chars ((a), left-hand side) and vitrinite-rich chars ((b, right-hand side).

Green = C, white = H, red = O, blue = N and yellow = S. ... 136 Figure 5-1. HRTEM results. Microphotograph for the inertinite-rich chars ((a), top left) with

scale bar = 5 nm. Lattice after image processing ((b), top right).

Parallelogram catenations ((c), bottom left). Distributions of aromatic raft lengths ((d), bottom right), where, dark and light bars = inertinite- and

vitrinite-rich chars, respectively. ... 159 Figure 5-2. The PAH char models used to determine the nucleophillic Fukui values of the

edge carbon sites, where (a), (b) and (c) = 3x3, 4x4 and 5x5 structures, respectively. ... 160 Figure 5-3. Geometry optimized structures for H-loss from (a) benzene molecule at (b)

para-, (c) meta- and (d) ortho-carbon sites. ... 165 Figure 5-4. Geometry optimised structures for H-loss from (a) 3x3 char structure at (b)-(f)

= Ct-, Cz-, Czi-, Cr1 and Cr2-carbons, respectively... 166

Figure 5-5. Structures with free edge carbon sites, where (a) = Benzene (left-hand side) and (b) = 3x3 char (right-hand side). ... 167 Figure 5-6. Various optimised structures for the CO2 chemisorption on single active sites

in the reactive C* edge. Top left is 3x3 char model (a), Top right is CO2

model (b), (c) is the 3x3 char model with free active sites and (d) to (h)

the chemisorption structures. Red = oxygen atoms... 169 Figure 5-7. Bond bending and elongations of geometry optimised CO2-C* chemisorption

structures, where: red = oxygen atoms. ... 172 Figure 5-8. Ground state reactants (a), (left-hand side) and products (b), (right-hand side)

(27)

xxvi

Figure 5-9. Transition state structure at Czi (a) (left-hand side). Ground state char(C=O)

quinone structure (b) (reactants) for the loss of second CO on Czi.

Numbers = bond lengths in Å. ... 176 Figure 5-10. Transition state structures for the decomposition reactions at respective

active sites (i.e. reactive edge), where, (a)-(e) = results at Czi, Ct, Cz, Cr1

and Cr2, respectively. Red = oxygen atoms. ... 178

Figure 5-11. Conversion-time plot for inertinite- and vitrinite-rich chars, where, (a), left-hand side = inertinite-rich char and (b), right-left-hand side = vitrinite-rich

chars. ... 180 Figure 5-12. Influence of isothermal reaction temperatures on the lumped reaction rates of

the chars. ... 180 Figure 5-13. Validation of actvation energy calculated according to atomistic reaction

kinetics with experimental results obtained from themogravimetric

experimentation. ... 181 Figure 5-14. Comparison of experimental and model predictions. ... 182 Figure 5-15. Comparison of activation energies. ... 183 Figure 5-16. Mechanisms of char gasification by CO2, where (a) = chemisorption of CO2 to

active sites in char, (b) = yield of first CO and (c) = yield of second CO. ... 185 Figure 6-1. Raw TGA results of the isothermal char-CO2 gasification at 900 ℃. ... 204

Figure 6-2. Influence of isothermal gasification reaction temperature on the reactivity of

the (a) IR (left-hand side) and (b) VR chars (right-hand side). ... 205 Figure 6-3. Comparison of the CO2 gasification reactivity of the IR and VR chars.

Conversion-time plot at 900 ℃ (a), left-hand side and 980 ℃ (b),

right-hand side, respectively. ... 206 Figure 6-4. Fractional carbon conversion vs reduced time plots of (a) Char IR; and (b)

(28)

xxvii

Figure 6-5. RPM fitting to the experimental data of (a) char IR, and (b) VR. ... 208 Figure 6-6. Arrhenius plots of the char-CO2 gasification reactions of chars IR and VR from

RPM data... 209 Figure A-1. Blocks prepared for petrographic analysis. ... 222 Figure A-2. Mayor macerals found during petrographic analysis. Inertodetrinite with

semifusinite ((a), top row left). Semifusinite ((b) top row middle). Micrinite and pyrite in colotellinite ((c), top row right). Colotellinite ((d), bottom row left). Copotellinite between inertinites ((e), bottom row middle).

Pseudovitrinite ((f), bottom row right). ... 223 Figure B-1. Diffractograms for inertinite- and vitrinite-rich particles on respective sets of axis.

Inertinite-rich particles ((a), top left) and vitrinite-Inertinite-rich particles ((b), top right). Zoomed (002) band for inertinite-rich particles ((c), bottom left) and vitrinite-rich particles ((d), bottom right). IR and VR = inertinite- and vitrinite-rich, respectively...228

Figure B-2. Comparison of XRD structural parameters. IR and VR = inertinite- and vitrinite-rich,

respectively. Numbers = temperature (℃). From top left to bottom right, (a) d002, (b) Lc,

(c) Nave, (d) La (10) and La (11), (e) fa, (f) Xa, (g) IOD, (h) IOD compared with

Xa...229

Figure B-3. Selected diffractogram patterns. Inertinite-rich chars ((a), top left) and vitrinite-rich chars ((b), top right) at 450, 700 and 1000 ℃. Reproducibility of diffractogram patterns for vtrinite-rich chars at 450 ℃ ((c), bottom left) and vitrinite-rich coals ((d), bottom right)...230

Figure C-1. The HRTEM image for the vitrinite-rich chars generated at 1000 ℃. Vitri1.G = area to be cropped ((Step 1), left-hand side) and cropped area ((Step 2), right-hand side)...231 Figure C-2. Fast Fourier transform forward ((Step 3), left-hand side) followed by a general ideal filter ((Step 4), right-hand side)...231 Figure C-3. Inverse fast Fourier transform filter ((Step 5), left-hand side) followed by a suitable threshold

level ((Step 6), right-hand side)...232 Figure C-4. Binary image inverted ((Step 7, left-hand side) followed by an ideal conditional smoothing five times at high median ((Step 8, right-hand side)...232

(29)

xxviii

Figure C-5. Skeletonised binary image ((Step 9), left-hand side) followed by trimming ((Step 10, right-hand side)...233 Figure C-6. Final results of lattice coloured by length ((Step 11), left-hand side) and representation of all selected features properties after a calculation Step.12 (right-hand side)...233 Figure C-7. Scale adjustment...234 Figure C-8. Example of unwanted features extracted during the trimming process...235 Figure E-1. Reproducibility results for char at 960 ℃...237

(30)

xxix

List of tables

Table 1-1. Scope of the research project... 8 Table 2-1. Main coal gasification reactions ... 21 Table 2-2. Four stages for the change of the micro-texture during carbonization up to 3000

℃ (Feng et al., 2002:481). ... 35 Table 2-3. Typical structural parameters from solid-state 13C NMR for vitrinite-rich

Waterberg and inertinite-rich Highveld coals (Van Niekerk et al.,

2008:290). ... 38 Table 3-1. Proximate and ultimate analysis of parent coals and chars. ... 87 Table 3-2. Major physical and structural parameters of demineralised coals and de-ashed

chars. ... 88 Table 3-3. Vitrinite random reflectance for the density separated (parent) coals. ... 89 Table 3-4. Maceral compositions for the parent coals (% by volume). ... 90 Table 3-5. Microlithotype analysis of the parent coals. ... 91 Table 3-6. Char form analysis and summary of major reflectance properties, following

classification published in Everson et al. (Everson et al., 2008:3082). ... 94 Table 3-7. Structural and lattice parameters for inertinite and vitrinite coals and chars. ... 98 Table 3-8. Summary of structural parameters extracted from the XRD diffractograms. ... 101 Table 4-1. Proximate and ultimate analysis of parent coal and char ... 121 Table 4-2. Summary of structural parameters extracted from the XRD diffractograms ... 125 Table 4-3. Allocation of parallelogram-shaped aromatic fringes from the HRTEM fringe

results ... 130 Table 4-4. Comparison between the inertinite- and vitrinite-rich char models ... 137

(31)

xxx

Table 4-5. Model and experimental results for the inertinite- and vitrinite-rich chars with C normalised to 1000 ... 138 Table 5-1. Definition of H-terminated edge carbons, including position and quantity. ... 160 Table 5-2. Reactivity of 3x3 ((a), left-hand side), 4x4 ((b), middle) and 5x5 ((b), right-hand

side) coal char models from f+(r). ... 162 Table 5-3. Calculated ∆Eb and Ereac values for the loss of H from the benzene and 3x3

char structures. ... 167 Table 5-4. Distribution of Fukui indices on the 3x3 char model with a reactive edge. ... 169 Table 5-5. Simulation energetics for CO2-char chemisorption at the reactive edge. ... 171

Table 5-6. Computed Ereac and ∆Eb for loss of the first CO gas molecule. ... 174

Table 5-7. Computed Eb and Ereac for loss of second CO gas molecule. ... 177

Table 5-8. Estimated reaction parameters. ... 182 Table 6-1. Reaction conditions for char-CO2 gasification experiments. ... 203

Table 6-2. Model parameters. ... 207 Table 6-3. Models used, structural parameter, and kinetic parameters obtained for

char-CO2 gasification reaction by other investigators. ... 210

Table A-1. Theoretical standard deviation and repeatability limit of the percentage of a

component, based on counts of 500 points. ... 224 Table A-2. Classification system used in the petrographic analyses of chars. ... 225 Table D-1. Spin states for the molecular structures used in Chapter 5.. ... 236 Table F-1. Determination of experimental error from different Arrhenius plots (N = 7; DF = 14; CI = 95%)*...238

(32)

xxxi

Table F-2. Summary of estimated activation energies from the gasification experiments with uncertainties...238

(33)

xxxii

Publications in peer-reviewed journals

This section contains published work in accredited, peer-reviewed journals.

Publication 1. Mokone J. Roberts, Raymond C Everson, Hein W.J.P. Neomagus,

Gregory N. Okolo, Daniel Van Niekerk, Jonathan P. Mathews. The

characterisation of slow-heated inertinite- and vitrinite-rich pyrolysis chars from selected South African coalfields. FUEL, 2015. 158: p.

591-601.

Publication #2. Mokone J. Roberts, Raymond C. Everson, Hein W.J.P. Neomagus,

Daniel Van Niekerk, Jonathan P. Mathews, David J. Branken.

Influence of maceral composition on the structure, properties and behaviour of chars derived from South African coals. FUEL, 2015.

142: p. 9-20.

Publication #3. Mokone J. Roberts, Raymond C. Everson, George Domazetis, Hein

W.J.P. Neomagus, Cornelia G.C.E. Van Sittert, Gregory N. Okolo, J.M. Jones, Daniel Van Niekerk, Jonathan P. Mathews. Density

functional theory molecular modelling and experimental particle

kinetics for CO2-char gasification. CARBON, 2015. 93: p. 295-314.

Publication #4. Gregory N. Okolo, Raymond C. Everson, Hein W.J.P. Neomagus,

Mokone J. Roberts, Richard Sakurovs. Comparing the porosity and

surface areas of coal as measured by gas adsorption, mercury intrusion and SAXS techniques. FUEL, 2015. 141 (0): p. 293-304.

Publication #5. Gregory N. Okolo, Raymond C. Everson, Hein W.J.P. Neomagus,

Mokone J. Roberts, Richard Sakurovs. Chemical-structural

properties of South African bituminous coals: Insights from wide

angle XRD-carbon fraction analysis, ATR-FTIR, solid state 13C

(34)

xxxiii

Peer-reviewed conference proceedings

Conference #1. Mokone J. Roberts, Raymond C. Everson, Hein W.J.P. Neomagus.

2012. The molecular structure of selected South African coal-chars

to elucidate fundamental principles of coal gasification. Centre for

High Performance Computing. National Meeting and Conference. International Convention Centre, Durban, KwaZulu-Natal Province, South Africa.

Conference #2. Mokone J. Roberts, Raymond C. Everson, Hein W.J.P. Neomagus.

2012. Advanced characterisation of chars derived from inertinite-

and vitrinite-enriched coal towards molecular structures. The South

African Institution of Chemical Engineers. Champagne Sports Resort and Conferencing Venue. Central Drakensberg, KwaZulu-Natal Province, South Africa.

Conference #3. Mokone J. Roberts, Raymond C. Everson, Hein W. J. P. Neomagus,

Jonathan P. Mathews, George Domazetis, Cornelia G.C.E. van Sittert. 2013. Molecular mechanics to model coal char structures

and DFT to model their reactivity with CO2 gas for synthetic gas

production. Centre for High Performance Computing. National

Meeting and Conference. International Convention Centre, Cape Town, Western Cape Province, South Africa.

Conference #4. Mokone J. Roberts, Raymond C. Everson, Cornelia G.C.E. Van Sittert,

Hein W.J.P. Neomagus, Daniel Van Niekerk, Jonathan P. Mathews, George Domazetis. 2013. The molecular structure of selected

South African coal-chars to elucidate fundamental principles of the reaction between char and carbon dioxide. International

Conference of Coal Science and Technology. State College City, Pennsylvania State, USA.

(35)

xxxiv

Conference #5. Mokone J. Roberts, Raymond C. Everson, Hein W. J. P. Neomagus,

Daniel Van Niekerk, Jonathan P. Mathews, George Domazetis, Cornelia G.C.E. Van Sittert. 2013. Properties of chars derived from

South African inertinite- and vitrinite-rich coals using molecular modelling. Fossil Fuel Foundation. Stonehenge Conference Centre,

(36)

1

Chapter 1: General introduction

This thesis comprises a collection of three papers prepared for publication in accredited journals. The three papers respectively cover the characterisation of coal chars, construction of large-scale molecular representations of coal chars and atomistic reactivity simulation of the reactivity of coal char models with carbon dioxide, including validation of results with experimental work. Chapter 1 gives a brief introductory overview and orientation of this thesis from the beginning, through the execution steps and, finally, the conclusion. Accordingly, the chapter is subdivided into sections 1.2 to 1.6 with specific attention to the background behind the study, attributes of the project motivation, the specific problem in the current knowledge on the topic and opportunities for the research at hand, formulated hypothesis of the central theoretical statement, research objectives and project scope respectively.

1.1 Background

Coal plays an important role towards the world’s primary sources of energy and is likely to continue for foreseeable time. Coal has spurred the industrial revolutions of many countries because a variety of chemicals and compounds can be derived from coal and it has also been an important contributor to infrastructure for development. For example, in China, coal is the major fossil fuel consumed, facilitating the growing economic development. South Africa makes electricity and liquid transportation fuels from coal (DMR, 2011/2012:201). Many industrialised nations, such as the United States, have for decades selected coal as the fuel of choice (Smith, 1994). Other industrial spin offs by coal is the metallurgical coke, which is vital in the production of steel.

Coal contributed 27.3% of world primary energy in 2012 (EIA, 2012). Total world’s coal reserves were estimated at about 860.9 Bt in 2012, according to the mineral economics statistics collated by the South African Department of Mineral Resources (DMR, 2011/2012:201). South Africa holds the world's eighth-largest amount of recoverable

(37)

2 coal reserves (30.2 billion tons). This value accounted for 95% of total African coal reserves and nearly 4% of total world reserves. Coal producing countries of the world aggregated about 7695.5 Mt in 2012 and South Africa was a significant participant, accounting for the 7th highest production. In the same year, China alone accounted for more than three-quarters of incremental coal production, while domestic consumption was more than three times that of global trade (IEA, 2012). The top six global coal suppliers include Indonesia, Australia, Russia, USA, Colombia and South Africa (DMR, 2011/2012:201).

Coal is highly heterogeneous and requires several analytical techniques for its characterisation to predict its behaviour during conversion processes such as combustion and gasification. Standard analyses such as proximate, ultimate and ash provide only bulk characteristics (Gupta, 2007:451). The utilisation behaviour of coal based on these analyses does not adequately describe the impact of coal quality on conversion efficiencies and plant performance. The heterogeneity of the organic material in the coal arises from maceral constituents, classified in three groups, namely, inertinite, vitrinite, and liptinite (Van Krevelen, 1993; Falcon, 1987:323; Falcon, 1986; Yu

et al., 2003:1160). The mineral and volatile matter contents, functional groups and

heterogeneity of coal macerals causes wide diversity in coal particle behaviour during pyrolysis, combustion and gasification processes and contributes to the complexity of structure of coal and its chars (Yu et al., 2003:1160). Consequently, in-depth studies are requited. A number of analytical techniques, such as XRD, 13C NMR and HRTEM, are available to provide information on the chemical structure of coal.

1.2 Motivation

Advances in the knowledge of chemical structure of coal led to more than one hundred proposed molecular level representations (models) of coal during the past 50-100 years (Mathews and Chaffee, 2012:1). Some of these models were enhanced by the development of high performance computational techniques. The chemical information

(38)

3 was based almost exclusively on the carboniferous coals from the northern hemisphere. In contrast to this, there are only two molecular models based on the inertinite- and vitrinite-rich coals from the southern hemisphere (Niekerk and Mathews, 2010:73). The most economic utilisation of coal involves combustion and gasification processes, which consist of a char generation step prior to reaction with oxygen/carbon dioxide/steam. The knowledge of char structure and its impact on reactivity is desirable to efficient operation of combustion and gasification processes. Therefore, to supplement the scarce molecular representations of the Gondwana coals published by Van Niekerk et

al. (Niekerk and Mathews, 2010:73), a project was formulated to construct molecular

representations of chars derived from the South African inertinite- and vitrinite-rich coals. The resulting atomistic structures could be used to evaluate gasification reaction mechanisms with CO2 using molecular modelling techniques.

1.3 Problem statement

Several papers exist on the molecular modelling studies to evaluate the reactivity of oxygen and oxygen-carrying gases with carbonaceous material (Montoya et al., 2002:4236; Montoya et al., 2003:29; Sendt and Haynes, 2005:2141; Sendt and Haynes, 2011:1851; Sendt and Haynes, 2005:629; Radovic, 2005:907; Radovic, 2009:17166). The studies include the chemisorption of oxygen-carrying gases on often simplistic representations of carbonaceous surfaces using the DFT modelling techniques (Montoya et al., 2003:29; Sendt and Haynes, 2005:2141; Sendt and Haynes, 2005:629; Radovic, 2005:907). Montoya et al. used ab initio methods to model the kinetics of elementary reaction in carbonaceous material (Montoya et al., 2002:4236). Radovic et

al. used DFT to model the active carbon sites responsible for reaction with the

oxygen-carrying gases and to demonstrate a dissociative chemisorption reaction of CO2 on the

active carbon sites (Radovic, 2009:17166; Radović et al., 1983:849; Radovic, 2005:907). However, investigations were often on the chemisorption, dissociation or decomposition reactions on the carbonaceous surfaces, respectively, but not on all three reactions within a single investigation, which may represent a complete

(39)

4 gasification reaction mechanism. In addition to this, studies adopted carbon models not necessarily derived from coal. Coal macerals impart structural differences, consequently reactivity differences (Sun et al., 2003:669; Zilm et al., 1981:717). Fletcher et al. also highlighted that “the similarity in chemical structure of the chars is in contrast to the general trend in char reactivity (in oxygen)” (Fletcher et al., 1992:643). The authors concluded that the principal causes of differences in de-ashed char reactivity are physical structure (e.g., active surface area/porosity) rather than chemical structure (Fletcher et al., 1992:643).

Therefore, there was an opportunity to consider other properties such as crystallite size or molecular distributions and for the construction of molecular structures of chars derived from the inertinite- and vitrinite-rich coals and to evaluate the structure-reactivity relationship on an atomistic level. This relationship may be based on the char-carbon dioxide gasification reaction mechanism. Standard and other characterisation techniques could provide data, while high performance computing resources could enhance visualisation and accuracy.

1.4 Objective

1.4.1 Background

The research objectives essential for the attainment of milestones are outlined here and include the specific objectives.

The objective of this investigation is to construct molecular structures of chars derived from the inertinite- and vitrinite-rich South African coals. An attempt was made to evaluate the structure-reactivity relationship based on the CO2-char gasification reaction

mechanism at an atomistic level. The structural transformations of chars generated at 450, 700 and 1000 ℃ were analysed and the 1000 ℃ chars were targeted for the atomistic studies using high performance computational techniques. Coals with particle size distribution of 300-1000 μm were slowly heated at 20 ℃ minutes and kept at

(40)

5 respective temperatures for 60 min. Vapour phase secondary reactions of primary volatiles were minimised by inert gas flow rate of 1.5 dm3 min-1 (Xu and Tomita, 1987:627; Ahuja et al., 1996:272).

1.4.2 Specific objectives

Four specific objectives are outlined below as essential pillars of the investigation.

1.4.2.1 Characterisation

Standard analyses. These include chemical and physical parameters by means of

petrographic, proximate, ultimate, skeletal density, surface area and porosity analyses of density-separated inertinite- and vitrinite-rich coals and their corresponding chars generated at 450, 700, and 1000 ℃.

Other analyses. Structural characterisation techniques include the X-ray diffraction

(XRD) and solid-state 13C nuclear magnetic resonance (NMR) for the density separated inertinite- and vitrinite-rich coals and their corresponding chars generated at 450, 700, and 1000 ℃. The high resolution transmission electron microscopy (HRTEM) may only be used on the 1000 ℃ chars for the purposed of atomistic modelling.

1.4.2.2 Construction of molecular models

Analytical data was used to construct molecular models of the de-ashed chars generated from inertinite- and vitrinite-rich coals at 1000 ℃ using the available high-performance computational resources such as the Material StudioR. The results here were a set-up to explore the structure-reactivity relationship.

(41)

6

1.4.2.3 Perform atomistic reactivity simulations between char models and

carbon dioxide gas molecules

Atomistic simulations were conducted with a quantum mechanics technique of high accuracy, the density functional theory (DFT) for geometry optimisations, single-point energy calculations and transition state search calculations centred on proposed CO2

-char gasification reaction mechanism. This was meant as an attempt to demonstrate the utility of molecular models.

1.4.2.4 Determination of reaction kinetics

Reactivity measurements of de-ashed chars generated from inertinite- and vitrinite-rich coals at 1000 ℃ were carried out using an experimental thermogravimetric analyser with CO2 as reactant gas at atmospheric pressure. The results here were compared

with those from atomistic reactivity simulations.

1.5 Scope and outline of the research project

1.5.1 Background

The details of project execution including methods, results, conclusions, recommendations and contributions to the knowledge base of coal science and technology, are outlined in chapter dissections (Section 1.5.2), read in conjunction with Table 1-1.

The research commenced with the rationale for the selection of coals. The Permian-aged Gondwana coals in South Africa; of both the inertinite and vitrinite origin, have significant local and global economic impact. However, the volume of research on these coals is far too low compared to their carboniferous counterparts in the northern hemisphere. Therefore, the two coals from the Witbank #4 and the Waterberg Upper Ecca seams were selected. The coal samples were subjected to laboratory density

(42)

7 separation techniques to obtain inertinite and vitrinite-rich aliquots, followed by different characterisation techniques. The characterisation was guided by the coal petrography to first verify the suitability of the coals with respect to rank and maceral properties, including their transitions during heating; and to assess opportunities for project expansions. Low (450 ℃) and medium-to-high (700-1000 ℃) temperatures were selected for this study. The transitions of coal properties at (700-1000 ℃) temperatures drew enormous attention and culminated in the study of atomistic properties of the 1000 ℃ chars that may give hope for a detailed structure-reactivity relationship. The particle size distribution of chars was 300-1000 μm. It is postulated that this particle size distribution and the heat treatment temperatures of 700-1000 ℃ are typical in fluidised bed processes. Computational techniques of highest accuracy, namely, density functional theory (DFT), were used. Efforts were made to validate the DFT results with experimental methods and valuable inferences were made. Milestones were considered for publication/ presentation in accredited journals/conferences.

1.5.2 Breakdown by chapter

Chapter 1. Contains the background behind the study, project motivation, the specific

problem in the current knowledge on the topic and opportunities for the research at hand, formulated hypothesis of the central theoretical statement, research objectives and project scope. The project scope included the background and project breakdown by chapter.

(43)

8 Table 1-1. Scope of the research project

Witbank #4 seam Coal Waterberg Upper

Density separated Density separated

Inertinite-rich coal vitrinite-rich coal

Charring Charring

450 ℃ 700 ℃ 1000 ℃ 450 ℃ 700 ℃ 1000 ℃

De-ashed chars Demineralised coal De-ashed chars

Characterisation Characterisation Characterisation

Chapter 3 Pulication 1 Chapter 3

Large-scale Molecular modelling With Molecular mechanics Large-scale Molecular modelling

Chaper 4 Publication 2 Chapter 4

Fringes of highest frequency Fringes of highest frequency

Obtained reaction enthalpy and activation energy

Comparison

Chapers 5 and 6 Publication 3 Chapters 5 and 6

Conclusions

Atomistic reactivity modelling Atomistic reactivity modelling

Aromatic fringes by parallelogram catenations method

Chemically controlled CO2-char gasification expenriments Atomistic modelling of CO2-char

reaction mechanism Identified reactive carbon edge

Chemically controlled CO2-char gasification expenriments

With quantum mechanics (DFT)

(44)

9

Chapter 2. Comprises an overview of the most relevant volume of knowledge in

literature citations on the four core research activities, coal/char characterisation, large-scale molecular modelling of coal chars, CO2-char gasification and atomistic-scale

gasification (reactivity) simulations using DFT. This is to complement literature given in chapters.

Chapter 3. This is composed of coal/char characterisation techniques. These entail

physical and chemical structural transformations of inertinite- and vitrinite-rich coals at at 450, 700, and 1000 ℃. Correlations among the results are also highlighted. HRTEM technique only applied to the 1000 ℃ chars derived from inertinite and vitrinite-rich coals to construct large-scale molecular models and study structure-reactivity relationship.

Chapter 4. Detailed construction and evaluation of large-scale molecular models of

chars derived from inertinite- and vitrinite-rich coals at 1000 ℃ is presented in this chapter.

Chapter 5. Atomistic/molecular reactivity modelling of CO2-char gasification using DFT

technique, including the parameters for geometry optimisation, single-point energy calculations and transition state theory.

Chapter 6. Description of the experimental apparatus (TGA) used for the gasification

experiments, including details of procedures and experimental programme. Special attention here is given to experimental gasification results of chars derived from inertinite- and vitrinite-rich coals at 1000 ℃.

Chapter 7. General conclusions for this study are drawn and include an account of

results from this investigation, contributions to the knowledge base of coal science and technology and recommendations for future research.

Appendix. Additional information on petrography, XRD, HRTEM, DFT and particle

(45)

10

Nomenclature

Acronym Definition

Bt Billion tons

DFT Density functional theory

DMR Department of Minerals Resources (South Africa) EIA Energy Information Administration

HRTEM High resolution transmission electron microscopy ss 13C NMR Solid state nuclear magnetic resonance spectroscopy XRD X-ray diffraction

The molecular structure of selected South African coal-chars to elucidate fundamental principles of coal gasification